Proceedings of the 1st Rail Transportation Division Fall Technical Conference

RTDF2007
September 11-12, 2007, Chicago, Illinois, USA

RTDF2007-46015
DEVELOPING STRATEGIES FOR MAINTAINING TANK CAR INTEGRITY DURING
TRAIN ACCIDENTS
David Tyrell
Karina Jacobsen
Brandon Talamini
Michael Carolan
Volpe National Transportation Systems Center
Cambridge, Massachusetts, USA
ABSTRACT
Accidents that lead to rupture of tank cars carrying
hazardous materials can cause serious public safety hazards and
substantial economic losses. The desirability of improved tank
car designs that are better equipped to keep the commodity
contained during impacts is clear. This paper describes a
framework for developing strategies to maintain the structural
integrity of tank cars during accidents.
The target of this effort is to design a tank car capable of
surviving impacts without loss of lading at twice the impact
speed of current equipment (or, equivalently, is capable of
absorbing four times the impact energy). The methodology
developed breaks down the process into three steps:
1. Define the impact scenarios of concern
2. Choose strategies to mitigate failure modes
present in each scenario
3. Design and select technology and tactics to
implement the mitigation strategies
The railroad accidents involving tank cars that occurred in
Minot, ND, in 2002, and Graniteville, SC, in 2005, are
examined to define the impact scenarios. Analysis of these
accidents shows that two car-to-car impact scenarios are of
greatest concern: head impact, where railroad equipment
impacts the end of a tank car and possibly overrides it, and
shell impact, where the tank car is impacted on its side,
possibly off center.
A conceptual design that can protect its lading at twice the
impact speed of current equipment in the car-to-car impact
scenarios is being developed. The conceptual design includes
four functions to meet the impact requirements: blunts the
impact loads, absorbs collision energy, strengthens the tank,
and controls the load path to assure that loads are blunted and
that energy is absorbed before the tank is loaded.

Preliminary studies of available weight and space,

strategies for increasing energy absorption, and strategies for
strengthening the head and shell are ongoing; this paper
describes the current results of these studies. Additional studies
are also ongoing. The steps required to complete this effort are
also described.
INTRODUCTION
In support of the Federal Railroad Administration (FRA),
the Volpe National Transportation Systems Center (Volpe
Center) is conducting research on improving tank car safety.
The results of this research are being used to help form the
technical basis for ongoing tank car safety rulemaking, which is
being carried out by the Pipeline and Hazardous Material
Safety Administration (PHMSA) and FRA. This rulemaking
activity is focused on tank cars which carry materials that are a
Toxic Inhalation Hazard (TIH), such as chlorine. This research
is intended to support the activities described in a press release
issued by FRA on January 16, 2007 [1]:
Our goal is to jump beyond incremental design changes,
[FRA Administrator Joseph H.] Boardman said. We and our
partners are looking to apply the latest research and advanced
technology to provide increased safety for rail shipments
posing the greatest safety risk, he explained, noting that FRA
is considering issuing new, more robust Federal design
standards for hazardous materials tank cars and hopes to issue
a final rule in 2008.
The goal set for this research is to double the car-to-car
impact speed for which integrity of the tank can be maintained
and the commodity contained. This paper describes one aspect
of the tank car safety researchthe development of an
improved conceptual design. The purpose of the conceptual
design is to show that the chosen goal can be met. Results
from the development of the improved conceptual design are

This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
1

being used to estimate benefits and costs, demonstrate

techniques for evaluating compliance with selected
requirements, and exemplify technologies needed to increase
tank integrity in accidents. Companion papers describe other
aspects of this research, including full-scale testing of the
baseline equipment [2], material failure modeling [3], train
collision dynamics modeling [4], and an overview of the entire
research effort in support of the PHMSA and FRA rulemaking
activity [5]. While significant progress toward the goal has
been made, the research is not yet completed.
APPROACH
The approach used in conducting this research is illustrated
in the flow diagram shown in Figure 1.
1. Define the collision scenarios of concern. The
accidents identify the conditions for which the
commodity is to be contained. These conditions
include the collision or derailment speed, the train and
other objects, and the track conditions [4, 6]. The goal
in developing the collision scenarios of concern is to
develop a limited number of scenarios that bound the
range of conditions leading to loss of lading. These
scenarios are used as the basis for evaluating the
effectiveness of alternative tank car designs.
2a. Develop information on the features of existing
designs that influence crashworthiness. Information on
the design details of the equipmentthe geometry of
the structure, material properties, welding and
attachment detailsare developed for use in analytic
models and in the development of test articles [7, 8].
2b. Develop potentially improved designs. A clean-sheet
design is being developed. The functions and features
of the clean-sheet design are also being evaluated to
determine if they can be incorporated into existing
designs. This paper focuses on this aspect of the
research.
3. Evaluate the effectiveness of existing design and
potentially improved design equipment. From postaccident results, it can be seen how effective the
equipment was in containing the commodity. Gaps
and uncertainties exist in the information available
from accidents. Analyses and tests are used to fill in
these gaps [2, 3, 5, 6, 7, and 8]. Analytic models and
tests, similar to those developed and conducted for the
conventional equipment, are used to evaluate the
effectiveness of the alternative designs.
4. Compare the effectiveness of the alternative designs
with the existing designs. For a given accident
scenario of concern, comparisons are made in terms of
the maximum collision speed for which the
commodity would be contained.
5. In order to evaluate the crashworthiness of existing
and alternative designs, analytic tools and testing
techniques are developed and refined. Accidents
provide information for comparison to analytic model

predictions and help provide some level of assurance

of the fitness of the models. Much is often unknown
about an accident, such as the precise collision speed
and exact initial conditions for the equipment. Testing
provides more detailed information for comparison
with analysis predictions [2].
5
2a

Existing
Equipment
Designs

Develop
Scenarios
2b

Potentially
Improved
Designs

Modeling
and
Testing

Evaluate

4 Compare
Effectiveness
of Potentially
Improved and
Existing
Designs

Revise

Figure 1. Flow Diagram, Tank Car Safety Research

Principal Tasks
This approach has been successfully applied to several
research projects in the area of rail equipment crashworthiness.
Examples include development of crush zones (referred to as
crash energy management or CEM) for passenger equipment
[9, 10, 11], a state-of-the-art cab car end frame [12] for
preserving the operators space, an improved workstation table
for limiting abdominal injury [13], and an optimized commuter
seat design [14] for protecting both rear and forward facing
passengers. Concepts developed with this approach include
deformable locomotive short hood and collision posts [15] for
preserving the operators space and inflatable structures for
locomotive operator protection [16].
IMPROVED DESIGN DEVELOPMENT
Collision dynamics and structural response, as well as
material failure are the tools used in the design development.
The collision dynamics determine how the loads are applied to
the structure. The structural response, in combination with the
collision dynamics, determines the magnitude of the loads. For
a given impact speed, if the structure is relatively soft, then the
impact loads will remain low; if the structure is relatively stiff,
then the loads may become large. The magnitude of the load in
combination with the structural response and material
properties determines if material failure occurs. For a given
material and load, material failure may occur with an impacting
object that focuses the load on the tank, while material failure
may not occur for an impacting object that disperses the load
on the tank [3].
Figure 2 shows a flowchart of the process for developing
the improved tank car design. The design requirements
describe the desired performance of the tank car in detail.
Using these, concepts for meeting the requirements were
generated, rated, ranked, and selected. Preliminary design

This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
2

studies to support the refinement of the selected concept are

ongoing. This paper describes the selected design concept, as
well as results from preliminary design studies.
Design Goals

Service, Manufacturing,

Crashworthiness,
etc.

Generation of
Alternative
Concepts
Identify and
organize
concepts

Rating and
Ranking of
Concepts
Assess
concepts
potential for
meeting
requirements

Concept
Selection

Preliminary
Design
Studies

Choose
concept with
most potential

Develop
concept details

at the tanks centerline. The improved tank car must be able to

retain its lading for an impact with four times the energy as the
impact for which the baseline tank car can retain its lading. As
for the head impact design requirement, meeting this goal
would double the impact speed for which the integrity of the
tank can be maintained. As described in the references [2 and
3], the speed for which tank integrity is maintained depends on
the height and width of the nose of the blunt punch. The
relative increase in performance sought is the same for a wide
range of nose sizes.
Punch

X2 mph

Figure 2. Design Development Process

Ram Car

Key Design Requirements

The design requirements for all of the conventional
aspectsservice, manufacturing, inspection, test, and
maintenancedo not vary substantially from the current
requirements for conventional tank cars.
The greatest
difference is in the crashworthiness requirements.
Requirements have been added that prescribe how the tank car
must perform in selected impact conditions.
Figure 3 is a schematic of the head impact design
requirement. The tank car must be able to retain its lading for
an impact with a rigid fixed punch with a blunt end. The
impact occurs below the tanks centerline. The goal is to
increase the amount of impact energy for which the tank can
contain the commodity by a factor of four. Meeting this goal
would double the impact speed for which the integrity of the
tank can be maintained.
V

Punch

Tank Car

Plan View

Plan View

Tank Car

X2 mph

Ram Car

Tank Car

Elevation View

Figure 4. Schematic of Shell Impact Requirement

Figure 5 for the shell impact illustrates the connection
between speed and the energy the tank car must be able to
absorb without loosing its lading. The initial kinetic energy of
the moving equipment is translated into the work done to
elastically and plastically deform the structure. These impact
requirements drive the design differences between the
improved design and conventional design tank cars.

Tank Car

Elevation View

Figure 3. Schematic of Head Impact Requirement

Figure 4 is a schematic of the shell impact design
requirement. In this scenario, the side of the tank car is
impacted by a ram car with a blunt punch. The impact occurs

Figure 5. Initial Kinetic Energy and Energy Absorbed by

Tank Car in Shell Impact

This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
3

Absorb Collision Energy

The addition of an energy absorption component between
the shield and the tank can further increase the energy required
to rupture the tank. In essence, the energy absorption layer
decreases the speed of the impact experienced by the tank. The
energy absorption layer can significantly slow down the
impacting object before the tank is significantly loaded. Figure
7 illustrates the combined functioning of a shield and energy
absorption component.

Table 1. Conceptual Design Functions and Features

Figure 7. Schematic Illustration of Combined Head Shield

and Energy Absorption Component Function

Functions
Blunted impact
loads
Collision energy
absorbed
Strengthened tank
Controlled load
path to tank

Features
Sacrificial structure that
shields tank and absorbs
energy
Reinforcement of head and
shell
Carbody supports service
loads

Blunt Impact Load

Head shields on current cars principally act to blunt the
impact load. The shield makes the size of the impacting object
appear larger to the tank, spreading the load over a greater area,
and makes any sharp edges appear to have a larger radius,
dispersing the stress in the tank. By doing so, the energy
required to rupture the tank is increased. The concept is to
include a feature which can blunt the impact load on any part of
the tank, that is, to cover the entire tank with a shield. Figure 6
illustrates the functioning of such a shield.

Tank

Tank

Head Shield

Tank

Tank

Selected Concept
Aspects of the conceptual design include how the design is
intended to function and what features are included to perform
these functions. The functions are qualitative descriptions of
how the car structure is intended to behave in the selected
impact scenarios. The design features are intended to perform
the desired functions. The form is the realization of a design
with these features and functions, which would be a
constructible design. Several iterations of a constructible
design, with component tests and associated analyses, may be
necessary before a final production design is developed.
Table 1 lists the functions and associated features of the
selected concept. The conceptual design must perform the
following four functions to meet the impact requirements
associated with head and shell impact scenarios: blunt the
impact loads, absorb collision energy, strengthen the tank, and
control the load path to assure that loads are blunted and that
energy is absorbed before the tank is loaded.

Energy Absorbing Material

Strengthen Tank
Strengthening the tank allows an energy absorption
component to crush at a higher load. As suggested by Figure 5,
the higher the load required to deform the tank car (and its
components), the more energy required to rupture the tank car.
Since the head can snap through at a relatively low load [7],
the energy absorption crush load that can be supported without
reinforcement is relatively low. Similarly, the shell tends to
ovalize and dish at a relatively low load when impacted [2],
limiting the amount of energy that can be absorbed without
reinforcement.
If the tank remains the main service load bearing structure,
then the potential for weldments to lead to fatigue failures is of
concern. In such a case, bonding the reinforcement to head and
shell is a potential option. If the service load bearing structure
is separate from the commodity tank, and the tank does not
experience cyclic service loads, then the potential for fatigue
failure is much lower. In such a case, directly welding to the
head and shell may be a viable option.
Load Path
Accident history suggests that rupture can initiate
anywhere on the tank. Ruptures are more likely at certain
locations on the tankbelow the belt line and at the ends of the
car. However, cars appear to have been ruptured owing to
impacts near the draft sill/body bolster and on the top. These
accident results suggest that the entire car should be shielded
and that collision energy should be absorbed before the tank is
impacted. In order to accomplish this function, a separate
carbody structure supports the service loads.

Figure 6. Schematic Illustration of a Head Shield Spreading

Load and Dispersing Stress
This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
4

Integrated Design
Figure 8 shows a schematic of the features integrated into a
conceptual design. Potentially, these features could be applied
incrementally to existing designs, working from the inside out.
The tank could be strengthened with bonded stiffeners alone.
As described in the preliminary design study section of this
paper, such stiffeners have the potential to double the energy
required to rupture the tank. A sacrificial structure for blunting
the load and absorbing energy could be further added. The
tank stiffeners and sacrificial structure together could increase
the energy required to rupture the tank by a factor of four.
Such a car would be at least somewhat more vulnerable near
the draft sill/body bolster attachments, even if these
attachments act as structural fuses and fail in a prescribed
manner for prescribed loads. Impact loads that bypass the
sacrificial structure could be introduced to the tank through the
Achilles heel of the draft sill/body bolster. The external
carbody would eliminate this vulnerability and result in the
integrated conceptual design. For example, the continuous
center sill design relieves the tank from bearing the in-train
buff and draft forces and continues to be used in DOT
105A500W tank cars built for carbon dioxide service.
Reinforcement

Sacrificial Structure

Carbody

Strengthen Tank

Blunt Force and Absorb Energy

Carry Service Loads

preliminary design studies will be described in follow-on

papers once the studies have been completed.
Weight and Space
In order to achieve adequate performance of a new car
design, it may be necessary to change the shape and size of the
tank car. Such changes could result in issues related to how the
commodity is loaded and unloaded. Some facilities will have
to re-evaluate the top platform design in order to access the top
fittings. If multiple cars are loaded/unloaded at one time, then
car spacings where such fixtures are located will need to be
evaluated, perhaps requiring some to be redesigned.
Additionally, it is understood that the design must account for
placement of safety appliances. The details of the final car
layout will be defined in the next phase of design development
and discussed in future work.
The baseline car chosen for this study is a DOT 105J500W
car, the same class of tank car used in the full-scale crash tests
[2]. This baseline car is constructed to Association of
American Railroads (AAR) plate B clearance standards [17].
The clearance standard dictates the maximum car height,
width, and truck center spacing for compliant cars. For plate B,
the maximum allowable height of any portion of the car is 151 above top-of-rail. At the maximum allowable width of 108, truck centers cannot exceed 41-3. Figure 9 shows the
envelope for equipment built to plate B.

Design Integration
Service, Manufacturing, and Crashworthiness Goals

Figure 8. Schematic Illustration of the Integrated

Conceptual Design

15-1
10-8

41-3

(truck to truck)
PRELIMINARY DESIGN STUDIES
Preliminary studies of available weight and space,
strategies for increasing energy absorption, and strategies for
strengthening the head and shell are ongoing. Managed energy
Top of rail
absorption is necessary to achieve improvements in tank car
Figure 9. Maximum Dimensions for Plate B
crashworthiness. Weight and space are primary constraints on
energy absorption features. The collapse strength of the
For a car with the maximum distance between truck
support structure is also a constraint on energy absorption
centers, the length of structure outboard of the trucks must also
features. The stronger the tank, the greater the load that the
be considered in the cars design. The swingout at the ends of
energy absorbing structure may crush at, and the smaller the
the car cannot exceed the swingout at the center of the car on a
crush distance required for a specified amount of energy
13o curve [17]. A car built to plate B with maximum distance
absorption.
between truck centers has a swingout at the center of the car of
This section describes the current results of the preliminary
5 . By extending the length of the car outboard of the trucks
design studies. Only the results for the head impact scenario
until this same swingout is achieved at the end of the car, a
are described. Parallel results have been achieved for the shell
maximum overall car length of 57-10 can be achieved.
impact scenario. Studies on additional design details of the
The baseline cars tank has an outer diameter of 8-6 3/10
sacrificial structures, including means of blunting the load and
and a nominal capacity of 17,300 gallons. This tank has a 1of the carbody structure, are also ongoing. The results of the
This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
5

7/8 clearance on either side before encountering the maximum

allowable car width. This clearance effectively limits the
maximum amount of energy absorbing material that can be
placed around the tank. As illustrated in Figure 5, increasing
the thickness of material being crushed at a given force will
result in more energy being absorbed.
Since the clearance diagram width cannot be extended, the
only way to include more space for energy absorbing material
around the tanks circumference is to decrease the tanks
diameter. Figure 10 shows a tank with a diameter 5 less than
the baseline overlaid on the baseline tank. The tank length has
been extended to maintain the same volume as the baseline
tank. With a head-to-head length of approximately 48,
sufficient clearance exists for placement of new structure
surrounding the head.

One concern associated with the addition of new structure

to the car is the additional weight that this will contribute. For
the improved crashworthiness tank car, the maximum allowable
weight is 286,000 lb. The new design includes many of the
same features of a conventional tank car, including the tank
assembly, truck assemblies, safety appliances, and an equal
lading capacity of 90 tons of chlorine. In addition, the new
design will include a strengthened tank, energy absorbing
material, and a structural frame to bear the service loads.
Table 2 shows the weights of components common to both
conventional and improved crashworthiness tank cars. These
preliminary values are given to estimate the amount of weight
that is available for new structure. Based on the weights of
existing tank car components and the 286,000 lb limit,
approximately 39,000 lb are available for new structure. This
available weight comes from increasing the car weight and
from incorporating the functions of the baseline cars jacket
and the stub sill/body assemblies into alternative components.

Bare Tank (t = 0.777")

Truck Assemblies (2)
Coupler/Draft Gear Assemblies (2)
Air Brake Assembly
Fittings and Housing Assembly
Manway
Ladders/Handholds/Railings
Handbrake Assembly
Paint/Markings

Weight (lbf)
38280
20580
2300
1310
1300
1300
1100
420
100

Chlorine

180000

Subtotal
Available for new structure

246690
39310

Target Energy Absorption and Kinematics

The crashworthiness performance of a tank car is defined
by its ability to maintain tank integrity for a given impact
condition at a minimum speed. Figure 5 illustrates the
relationship between the collision speed in the shell collision
scenario and the energy absorption requirement for a tank shell
to maintain tank integrity. For the baseline tank shown in
Figure 11, the shaded area under the force-indentation curve up
to the point of possible rupture represents the energy absorption
required for the head to maintain tank integrity. The tank car
research goal is to increase the collision speed for which tank
car integrity can be maintained by a factor of two. An
improved tank car that can withstand twice the speed will have
energy absorption requirements four times that of the baseline
tank car.
2500

2000

Force (kip)

Figure 10. Baseline Tank and Reduced Diameter Tank

Table 2. Weight Breakdown of Tank Car Components

Rupture
possible

1500

1000

500

Ebs ~ 1 million ft-lb

0
0

10

15

20

25

30

35

40

45

Dent (in)

Figure 11. Idealized Force-Indentation Characteristic for a

Conventional Head
Figure 12 illustrates the desired kinematic sequence of the
conceptual design tank car for the head impact requirement.
The conceptual design includes the crashworthiness features
described in Table 1 and illustrated in Figure 8. By designing a
tank car to crush in the prescribed sequence shown, the
collision unfolds in a more predictable and controlled manner.
For the head scenario, in state 1, initial contact is made between
the indenter and the outer tank carbody. In state 2, a sacrificial
energy absorbing layer begins to collapse. The footprint of the

This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
6

50

indenter is spread over a wider area; by state 3, a less severe

and more distributed load is applied to the commodity tank. In
state 4, the head deforms until rupture occurs in state 5.

State
1

Initial Punch Contact with Carbody

Energy Absorbed by Crushable

Structure

Crushable Structure Collapses,

Contact with Shell

Shell Crushes

Head Crushes

Head Ruptures

Shell Ruptures

State
1 Initial Punch Contact with Carbody

Absorbed by Crushable
2 Energy
Structure

Structure Collapses,
3 Crushable
Contact with Head

Figure 12. Head Impact Scenario: Improved Tank Car

Target Kinematic Sequence of Events
Figure 13 shows the energy absorption requirements of an
improved design. The desired force-indentation characteristic
was developed using the space constraints addressed earlier in
this paper and the features identified in Table 1. The resultant
curve represents an improved tank that has a sacrificial
structure that absorbs energy and distributes the load applied to
a strengthened head.

Figure 14. Shell Impact Scenario: Improved Tank Car

Target Kinematic Sequence of Events
Figure 15 shows force-indentation characteristics for the
baseline tank, a tank with an energy absorbing structure added,
and a reinforced tank with an energy absorbing structure.
Adding just an energy absorbing structure to a baseline shell
can potentially increase the energy for which tank integrity is
maintained by a factor of 2, increasing the speed for which tank
integrity is maintained by 50 percent. Further, strengthening
the shell and adding an energy absorbing layer potentially
increases the energy for which tank integrity is maintained by
four and doubles the speed for which tank integrity is
maintained.

Figure 13. Head Force-Indentation Characteristics

A desired kinematic sequence for the conceptual design
tank car in a shell impact and a target force-indentation
characteristic for the shell have also been developed. Figure 14
Figure 15. Shell Force-Indentation Characteristics
illustrates the shell kinematic sequence, and Figure 15 shows
the target force/indentation characteristic.
The desired
Design studies to determine how to achieve the desired
kinematic sequence for the shell impact parallels that for the
kinematics and the target force-indentation characteristics for
head impact. In state 1, initial contact is made between the
the head and shell are ongoing. If structures that meet these
indenter and the outer tank carbody. In state 2, a sacrificial
design goals can be developed, then an improved tank car can
energy absorbing layer begins to collapse. The footprint of the
meet the impact conditions illustrated in Figures 3 and 4.
indenter is spread over a wider area, and by state 3, a less
Head Strengthening
severe and more distributed load is applied to the commodity
As illustrated in the previous section, the capacity of the
tank. In state 4, the head deforms until rupture occurs in state
energy-absorbing components may be extended by increasing
5.
This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
7

the head. The energy required to deform the head is the

integral of the force-indentation curve. The area under the
force-indentation characteristic for the reinforced head is more
than twice the area under the characteristic for the baseline
head. Figure 18 also shows the force-indentation characteristic
for a head reinforced by thickening the material to 1.11 in. The
area under that curve is greater than the area under the baseline
curve; however, it is significantly less than the area under the
curve associated with the head reinforced with ribs.
2000

1500
Force (kip)

the strength of the commodity tank in its role as a support

structure. Various means of increasing the strength of the tank
are compared in this section.
Shell structures (such as aircraft fuselages and ship hulls)
are often reinforced with ribs; that is, with beams connected
along their length to the shell. This arrangement increases the
bending resistance of the shell with little additional weight.
Ribs may be attached by welding, glue bonding, riveting, or
they may be formed directly into the shell. Figure 16 illustrates
a schematic of spider web arrangement of ribs on a tank car
head. This arrangement is investigated with finite element
analysis with ribs that have a cross-sectional moment of inertia
of 12.6 in4 and a cross-sectional area of 12.6 in2. The head has
a thickness of 0.777 in.

Ribs (beams) attached

to surface of tank head

1000

500

Baseline
Thicker Head
Reinforced Head

0
0

10

15

20

Indentation (in)

Figure 16. Tank Head Reinforce with Spiderweb

Arrangement of Ribs
Figure 17 shows the deformed shape of the baseline head,
as well as the deformed shape of the head reinforced with ribs
for an impact in the center of the head with a blunt punch. The
figure shows the maximum indentation of both heads when
they have absorbed 1 million ft-lb of energy. The reinforced
head has deflected significantly less than the baseline head for
the same amount of absorbed energy.

Figure 18. Force/Indentation Characteristics, Baseline

Head, Head Reinforced with Ribs, and Thickened Head,
Center Impact with a Blunt Punch
Figure 19 shows the undeformed shaped of the reinforced
head, as well as the deformed shape after an impact from a
blunt punch at a location away from the center of the head.
This location was chosen to be between the reinforcements,
where they would likely be least effective. Nevertheless, the
deformed shape suggests that the reinforcements are indeed
still effective.

Baseline

Reinforced

Figure 17. Indentations of Baseline and Reinforced Head,

Center Impact with a Blunt Punch

Figure 19. Indentation of Reinforced Head, Offset Impact

with a Blunt Punch
Figure 20 shows the force-indentation characteristics for
the baseline and reinforced heads for an offset impact with a
blunt punch. The force levels are not as high as for the center
impact, for both the baseline and reinforced head. However,
the area under the force-indentation characteristic for the
reinforced head is again more than twice the area under the
characteristic for the baseline head.

Figure 18 shows the force-indentation characteristics for

the baseline and reinforced heads for an impact in the center of
This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
8

2000

Baseline
Reinforced

Force (kip)

1500
1000
500
0
0

10

15

20

Indentation (in)

Figure 20. Force/Indentation Characteristics, Baseline

Head and Head Reinforced with Ribs, Offset Impact with a
Blunt Punch
The head reinforced with ribs can also support an energy
absorption component that crushes at a higher load than the
baseline and thickened heads. Alternative means of absorbing
energy are currently being investigated, including plastic
foams, aluminum honeycomb, and steel sandwich structures.
This research suggests that a tank car reinforced with ribs and
such an energy absorbing structure could maintain tank
integrity for an impact with four times the energy as the
baseline tank car.
SUMMARY AND NEXT STEPS
Research is being conducted to support the rulemaking
activities of FRA related to the TIH tank cars. Research results
to date suggest that an improved design tank car can double the
car-to-car impact speed for which integrity of the tank is
maintained over that of conventional tank car designs.
Reinforcing the head alone can potentially increase the
maximum impact energy by a factor of 2, increasing the
maximum containment speed by 50 percent. Results suggest
that reinforcing the head and adding an energy absorbing layer
increases the tank integrity energy by four times and doubles
the maximum containment speed. While not presented in the
paper, similar results have been obtained for the impact
resistance of the shell.
The results of this research are being shared with the
Dow/Union Pacific/Union Tank Car Next Generation Railroad
Tank Car. Cooperative activities include the full-scale shell
impact testing of conventional equipment. As described in a
companion paper [5], FRA supported the Volpe Center in
designing these tests and simulating the results of the tests with
simplified models. Dow Chemical supported Applied Research
Associates Inc. in conducting simulations of the test using a
detailed model and supported the Transportation Technology
Center Inc (TTCI) in implementation of the tests.
While information is being shared, the design development
activities supported by Dow Chemical are separate and distinct

from the design development activities being conducted by the

Volpe Center in support of FRA. In the process chosen by
FRA, an overall strategy is evolved into a single concept with
required functions and features. These features are then refined
into detailed forms with preliminary design studies. This
approach has been used to develop crush zones for passenger
equipment [9, 10, and 11], the state-of-the-art cab car end
frame design [12], and improved locomotive crashworthiness
features [15]. Alternative design development approaches
exist; any of which may result in an improved tank car
conceptual design that differs from the conceptual design
described in this paper. The performance achievable with such
an alternative may exceed the performance associated with the
conceptual design described in this paper.
Preliminary design studies are ongoing for the further
refinement of the conceptual tank car design presented in this
paper, including ongoing studies of available weight and space,
strategies for increasing energy absorption, and strategies for
strengthening the head and shell. Studies on additional design
details of the sacrificial structures, including means of blunting
the load, and the carbody structure, are also ongoing.
In order to achieve the required performance, it is
necessary to change the size of the tank car. Such changes will
affect the how the commodity is loaded and unloaded. In
addition, it is understood that the design must account for
placement of safety appliances. The details of the final car
layout will be defined and discussed in future studies.
Component tests of a head strengthening strategy, as
discussed in this paper, are tentatively planned to assess the
results of analyses results for the mode shape and force/dent
characteristic. Depending on funding, component tests of shell
strengthening and the sacrificial structure(s) may also be
conducted.
ACKNOWLEDGMENTS
FRA Office of Research and Development sponsored the
work described in this paper. Ms. Claire Orth is the Chief of
the Equipment and Operating Practices Division.
Mr.
Francisco Gonzalez is the FRA program manager for research
on railroad tank cars.
Mr. Grady Cothen, Deputy Associate Administrator for
Safety Standards and Program Development, FRA, is leading
the effort to develop tank car safety standards. Mr. Eloy
Martinez, FRA program manager, is developing the technical
information needed to support the FRA and PHMSA
rulemaking effort. The authors thank Ms. Jo Strang, Associate
Administrator for Safety Enforcement, FRA, for her efforts to
coordinate the standards development and the tank car safety
research.
The authors also thank Ms. Michelle Priante and Mr. Philip
Mallon, Mechanical Engineers, Volpe Center, for their efforts
in developing concepts for improving tank car safety. Dr.
David Jeong, Senior Engineer, Volpe Center, is leading the
effort to evaluate the performance of baseline equipment.
Finally, the authors would like to thank Professor A. Benjamin

This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
9

Perlman, Senior Engineer, Volpe Center, for his assistance

through all aspects of this project.
REFERENCES
[1] Federal Railroad Administration, Development of New
Federal Design Standards for Hazardous Materials Tank
Cars to Benefit from Public-Private Partnership, U.S.
Department of Transportation, FRA02-07, Tuesday,
January 16, 2007.
[2] Tang, Y.H., Yu, H., Gordon, J.E., Priante, M., Jeong, D.Y.,
Tyrell, D.C., and Perlman, A.B., Analysis of Full-Scale
Tank Car Shell Impact Tests, Proceedings of the 2007
ASME Rail Transportation Division Fall Technical
Conference, September 11-12, 2007, Chicago, IL,
RTDF2007-46010.
[3] Yu, H., Jeong, D.Y., Gordon, J.E., and Tang, Y.H.,
Analysis of Impact Energy to Fracture Unnotched Charpy
Specimens Made from Railroad Tank Car Steel,
Proceedings of 2007 ASME Rail Transportation Division
Fall Technical Conference, September 11-12, 2007,
Chicago, IL, RTDF2007-460xx.
[4] Jeong, D.Y., Lyons, M.L., Orringer, O., and Perlman, A.B.,
Equations of Motion for Train Derailment Dynamics,
Proceedings of the 2007 ASME Rail Transportation
Division Fall Technical Conference, September 11-12,
2007, Chicago, IL, RTDF2007-46009.
[5] Tyrell, D.C., Jeong, D.Y., Jacobsen, K., and Martinez,
E.E., Improved Tank Car Safety Research, Proceedings
of the 2007 ASME Rail Transportation Division Fall
Technical Conference, September 11-12, 2007, Chicago,
IL, RTDF2007-46013.
[6] Paetsch, C.R., Perlman, A.B., and Jeong, D.Y., Dynamic
Simulation of Train Derailments, Proceedings of 2006
ASME International Mechanical Engineering Congress
and Exposition, November 5-10, 2006, IMECE200614607.
[7] Jeong, D.Y., Tang, Y.H., and Perlman, A.B., Engineering
Analyses of Railroad Tank Car Head Puncture,
Proceedings of 2006 ASME International Mechanical
Engineering Congress and Exposition, November 5-10,
2006, IMECE2006-13212.
[8] Coltman, M., and Hazel, M., Chlorine Tank Car Puncture
Resistance Evaluation, Volpe National Transportation
Systems Center, Cambridge, MA, Report No.
DOT/FRA/ORD-92/11, June 1992.
[9] Tyrell, D., Jacobsen, K., and Martinez, E., A Train-toTrain Impact Test of Crash Energy Management Passenger
Rail Equipment: Structural Results, American Society of
Mechanical Engineers, Paper No. IMECE2006-13597,
November 2006.
[10] Martinez, E., Tyrell, D., Rancatore, R., Stringfellow, R.,
and Amar, G., A Crush Zone Design for an Existing
Passenger Rail Cab Car, American Society of Mechanical
Engineers, Paper No. IMECE2005-82769, November
2005.

[11] Martinez, E., Tyrell, D., and Perlman, A.B., Development

of Crash Energy Management Designs for Existing
Passenger Rail Vehicles, American Society of Mechanical
Engineers, Paper No. IMECE2004-61601, November
2004.
[12]Mayville, R., Stringfellow, R., Johnson, K., and Tyrell, D.,
Rail Vehicle Car Cab Collision and Corner Post Designs
According to APTA S-034 Requirements, American
Society of Mechanical Engineers, Paper No. IMECE200344114, November 2003.
[13] Parent, D., Tyrell, D., Rancatore, R., and Perlman, A.B.,
Design of a Workstation Table with Improved
Crashworthiness Performance, American Society of
Mechanical Engineers, Paper No. IMECE2005-82779,
November 2005.
[14]Severson, K., Tyrell, D., and Rancatore, R.,
Crashworthiness Requirements for Commuter Rail
Passenger Seats, American Society of Mechanical
Engineers, Paper No. IMECE2005-82643, November
2005.
[15]Tyrell, D., Severson, K., Marquis, B., Martinez, E.,
Mayville, R., Rancatore, R., Stringfellow, R., Hammond,
R., and Perlman, A.B., Locomotive Crashworthiness
Design Modifications Study, Proceedings of the 1999
IEEE/ASME Joint Railroad Conference, Catalog Number
99CH36340, 1999.
[16] Zolock, J., Tyrell, D., Locomotive Cab Occupant
Protection, American Society of Mechanical Engineers,
Paper No. IMECE2003-44121, November 2003.
[17] Association of American Railroads, Technical Services
Division, Mechanical SectionManual of Standards and
Recommended Practices.

This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. Approved for
public release; distribution is unlimited.
10